Inspiration

We wanted to create something affordable and helpful. Although the internet did not prove too helpful with giving us an exact idea. While scrolling mindlessly through different product advertisement videos we came across magnetic steel ball drawing board. We wanted to create one that would change according to the letters perceived but we knew the magnetic metal balls would not hold up to touch. Some amount of researching brought us to push pull linear actuator motor electromagnets which would solve the case for us.

What it does

A document with printed text is aligned with the frame and the Raspberry Pi is switched on. The webcam captures the document and searches for text. If present, it uses Optical Character Recognition (OCR) and image processing to enhance the text and then extracts it. It is stored as a String. This String is then referred to one character at a time. This then refers to the built in text to braille dictionary to find its corresponding braille format. Each character ie, A-Z, 0-9, . , ! ? ; : - ‘ “ ” ( ) / (space) are denoted by the corresponding braille formats. Each braille character consists of a 3x2 array. The pins are numbered vertically from 1-6. Pins that have to rise are denoted by a 1 and the others are denoted by a 0. Therefore, B is denoted as [1,1,0,0,0,0] and C as [1,0,0,1,0,0] in the system. Using General Purpose Input Output (GPIO) pins of the raspberry Pi, it sends a voltage if the value of the pin is 1. Pins 16, 18, 22, 36, 37, 31 (GPIO 23, 24, 25, 16, 26, 6) represent pins 1, 2, 3, 4, 5, 6 of a braille cell respectively. These GPIO pins are connected to input side of a 6 channel relay. The output side of the relay has 3 output lines for each channel. Generally, each solenoid is connected to the ground. Hence they are all low. If a voltage is sent via the GPIO pin to the corresponding channel, the relay switches connection to a high voltage. This voltage is provided via a 9V power supply. Using 0.1 uF and 0.33 uF capacitors and 6 KA7805 voltage regulators connected in parallel, current is supplied to one wire of the solenoid while the other remain grounded. The solenoid is switched on and the corresponding pin rises in its position. After the specified time in the program elapses, the pins switch to the braille format of the next character. In case it is the last character, all pins fall back into the original position.

How I built it

After researching about push pull linear actuator, we started slowly searching the other hardware components it would need. We would need a raspberry Pi for the image processing, webcam for the image capture, capacitors and voltage regulators for the working of the solenoids. At first, we wrote the program to process the braille equivalent for an input string. Once that was done, we wrote the program for image processing then combined the two. Lastly, we build the hardware and connected the hardware and software components with each other.

Challenges I ran into

At first we tried to run the actuators on batteries. They drained very quickly and only provided enough current to power two of the solenoids. We brought some more voltage regulators and changed the power supply to a wired one by using an old mobile charger. While the solenoids worked, the image processing did not do too good of a job of recognizing the letters due to the camera having low resolution. Fortunately, we were able to find images with bolder lettering and a better webcam to help with the same.

Accomplishments that I'm proud of

This was one of our first projects, and we were able to do it myself and make it work. That itself is an accomplishment for us.

What I learned

We learned more about image processing, raspberry pi and its uses and also practical use of electrical components that we used in the project.

What's next for Image to Braille converter

We plan to advance the project further by adding in machine learning. Thus, the product can not only recognize printed text but also handwritten documents.

Built With

Share this project:

Updates